The present disclosure relates generally to failure prediction and health monitoring for semiconductor devices. Unanticipated failures in semiconductor devices cause system downtime in electrical systems such as utility grids and motor drives. While a semiconductor device may have an expected useable life, exposure to frequent thermal cycles or severe electrical stress will decrease the useable life of a semiconductor device. Device failure can be predicted by monitoring physical or electrical characteristics of a semiconductor, such as devices' ON-resistance (Rdson), gate leakage current (Igss), drain-source leakage current (Ldss), and device's parasitic capacitances (Coss, Ciss, Crss). Many of these parameters can be quantified and monitored by measurements of variation in voltage and current. Existing semiconductor monitoring systems suffer from a number of shortcomings and disadvantages. There remain unmet needs including increased reliability, decreased sensor complexity, and reduced costs. For instance, some monitoring systems use high bandwidth circuitry to measure fast changing electrical variables such as switch transition times, gate threshold voltage, parasitic capacitances, and inductances. In addition, for SiC semiconductor devices, measurements of certain parameters, such as the Rdson, are unfit for predicting device failure given their dependence on device temperature. There is a significant need for the unique apparatuses, methods, systems and techniques disclosed herein.